Association for Financial Markets in Europe. St. Michael s House 1 George Yard London EC3V 9DH. 24 August, 2012

Transcription

1 Submitted via to European Banking Authority Tower 42, Level Old Broad Street London EC2N 1HQ Dear Sir or Madam, Association for Financial Markets in Europe St. Michael s House 1 George Yard London EC3V 9DH 24 August, 2012 Consultation Paper on Draft Implementing Technical Standards on Supervisory Reporting Requirements for Leverage Ratio EBA/CP/2012/06 The Association for Financial Markets in Europe ( AFME ) 1 is pleased to submit the attached comments regarding the above Consultation Paper issued by the European Banking Authority (EBA) on 7 th June Thank you for soliciting our comments as part of your Consultation. We would be pleased to assist the EBA further if required. In particular, if you have any questions or desire additional information regarding any of the comments set out in our attached response please do not hesitate to contact the undersigned on or by at christine.brentani@afme.eu. Yours sincerely, Christine Brentani Managing Director Prudential Regulation Division, AFME Tel: +44 (0) Association for Financial Markets in Europe St. Michael s House, 1 George Yard, London, EC3V 9DH T: +44 (0) F: +44 (0) Company Registration No: Registered Office: St..Michael s House, 1 George Yard London EC3V 9DH

2 Association for Financial Markets in Europe Consultation Response 24 August 2012 EBA s Consultation paper on Draft Implementing Technical Standards on reporting requirements for leverage ratio (EBA/CP/2012/06) The Association for Financial Markets in Europe (AFME) welcomes the opportunity to comment on the European Banking Authority (EBA s) Consultation paper on Draft Implementing Technical Standards on supervisory reporting requirements for leverage ratio (EBA/CP/2012/06). AFME represents a broad array of European and global participants in the wholesale financial markets. Its members comprise pan-eu and global banks as well as key regional banks, brokers, law firms, investors and other financial market participants. AFME participates in a global alliance with the Securities Industry and Financial Markets Association (SIFMA) in the US, and the Asia Securities Industry and Financial Markets Association through the GFMA (Global Financial Markets Association). AFME is listed on the EU Register of Interest Representatives, registration number We summarise below our main issues for consideration with regards to the consultation, followed by answers to the individual questions raised. Issues for Consideration Implementation Timeline: It will be extremely difficult for institutions to comply with the 1 January 2013 implementation deadline for the leverage ratio reporting requirements given the delay in the CRD IV trialogue negotiations. The current proposals make no allowance for the extra time and cost that firms will face in order to implement the new CRD IV reporting requirements (which include all the COREP and FINREP requirements as well), nor the risk to data quality from the timelines envisaged. Also, it may be very difficult for regulators and the EBA to have the capability to receive and interpret the data that firms would provide, as they may not have the time to build and develop systems to receive and adequately interpret the data. Furthermore, extra time may be required if the EU Commission calls for any amendments to the final proposals which the EBA submits to them. Due to increasing uncertainty surrounding the finalisation of CRD IV, we recommend that this implementing technical standard become effective one year after the entry into force of CRD IV, and, at this stage, not before 1 January Given the amount of changes required, more time is necessary for institutions to effect the necessary implementation provided the templates and taxonomy are made available. We note the recent EBA s update on the finalisation of the standards on supervisory reporting and we recognise that the EBA has already delayed the reporting of Page 1 of 10

3 FINREP templates. We believe similar arguments for delaying the reporting of the leverage ratio reports apply. Firms do not have any precedent for providing a project of this scale of complexity in the short timeframes that is being proposed. Also, it is difficult for firms to progress without the answers to some detailed and technical questions for which answers are not yet available. This is delaying the build out of projects at institutions Level of Application: With regards to the scope of application, we would expect that the Leverage Ratio (LR) templates should be provided along the same legal entity lines as for the COREP template requirements. That being said, some clarity as to what is meant by consolidation at the COREP level is still required. This is especially an issue in countries where the COREP reporting is yet to be implemented. The requirement to report the LR on single entity, sub group and/or on a consolidated group basis adds complexity in the reporting that most companies existing IT systems are not able to support. The likely manual consolidation processes to the required sub group level can be time-consuming and complex, considering the need to net transactions between group entities that for other reporting purposes report on a standalone or fully consolidated group basis. We also support flexibility that allows adequate time for firms to complete sub group and group consolidated reports, during the parallel run period. Thus, we would recommend a grace period where institutions are required to only calculate the LR on a consolidated basis until the end of the calibration period. If the main purpose of this reporting is to assist the EBA in fine-tuning the calibration, a good sample would be provided on the basis of the consolidated calculations. Remittance Dates: AFME and its members also would recommend a phased leverage ratio reporting approach of allowing more time at the beginning, such as 90 days, then falling to 45 days and finally 30 working days. Otherwise, if data is provided before institutions are able to finalise their systems and controls, the data may not be of the desired quality. The remittance dates should be linked to when the Capital Requirements Regulation is finalised. There should be a grace period after the final version is published in the Official Journal followed by a phased approach for remittance of data. If institutions do not have enough time to establish reliable systems and approaches to gathering and verifying the required data, the quality of submissions will undoubtedly suffer. This may render any subsequent analysis of submitted data of less use than originally planned. At the minimum, AFME proposes a delay by one year for the mandatory use of the new reporting requirements until 1 January EBA Technical Support: The EBA will need to consider how it will be able to provide support for the many questions that institutions are likely to have during the initial implementation phase of the templates. It is also important that there is adequate support provided to institutions in order to clarify some of the definitions and reference points. AFME and its members have noted that the tight implementation period will require significant dialogue between the EBA and banks on the technical standards. The industry feels that the various issues, especially those relating to netting practices, require substantial technical support from the EBA between the conclusion of technical Page 2 of 10

4 standards and their implementation. Therefore, due to the short timeframe provided, we request the establishment of an EBA LR Support Team to provide guidance on any technical issues which firms face that will need policy interpretation. It would be useful for the EBA to have a help desk approach during the initial implementation phase at the minimum. Superfluous and Duplicative Data Requirements: The leverage ratio (LR) template requires firms to report data items that are not readily available within the existing regulatory or statutory data requirements, thus requiring banks to invest in new systems infrastructures and to potentially establish manual data collation processes. More specifically, the EBA LR template requests data that facilitates the calculation of the leverage ratio and, additionally, secondary data that is not directly linked to the ratio (for example on asset encumbrance). AFME and its members query the purpose of the additional data gathering requirements and we would be grateful to know what incremental systemic benefit can be achieved from gathering and interpreting this data. Also, there are a number of data items requested for the LR templates that are already being collected under COREP templates. This duplication of effort will undoubtedly add needless resource burden on firms. To summarise, we would like to know what the extra data beyond that required for the pure LR calculations will be used for. Also, we would ask that consideration be given to minimising the duplication of data requests for the LR templates if the data is already requested in the COREP templates. Once the CRR text is finalised, the EBA should conduct a comprehensive review of all supervisory reporting requirements to eliminate any duplication of similar data item collection and make available a common data definition in case the same data ought to remain in different reporting templates. Only then will it be appropriate to finalise this implementing technical standards (ITS). (We fully support the recently announced workshop which the EBA will hold with industry to discuss technical issues of the ITS 1 ). Additionally, for data which is superfluous to the LR calculation, if this data is ultimately required, it should be provided on a quarterly basis at a minimum. Unsettled securities purchases and sales (Other Assets): We note that Article 416, in conjunction with Article 106(1), requires firms to include these exposures in the leverage ratio calculation at their balance sheet value. Some local accounting standards require such purchases and sales to be reported net on the balance sheet, whereas others require reporting at a gross level. Reporting at a gross level will clearly result in different leverage ratio outcome than reporting at a net level. This gives rise to non-level playing field issues across jurisdictions, and also has additional relevance for firms where their local and Head Office accounting standards differ in this area. The impact of this on firms leverage ratios can be significant. We suggest that reporting at a net level is the most straightforward way to ensure a level playing field outcome for these short-term items. 1 Implementing Technical Standard Page 3 of 10

5 Responses to Questions Q1: Do institutions agree with the use of existing and prudential measures? Is there additional ways to alleviate the implementation burden? There are several additional data items in the LR templates that are not required specifically for the LR calculations and their purpose is unclear and requires clarification. For example, the request for asset encumbrance data in template LR8 is a complex area which could result in a particularly onerous requirement for firms. Also, more guidance is needed with regards to the netting requirements. Furthermore, some of the additional data can be derived from other COREP reporting. As an example, the LR6 data corresponds to counterparty credit risk data and, apart from the classification differences, are duplicated reporting items (in fact, for IRB banks, some of the requested breakdowns do not exist). We are unclear as to why the data needs to be reported twice. Ensuring that the data point models (DPM) for both reporting purposes are aligned would minimise the need for additional data in the leverage ratio template and would reduce the burden on firms to provide similar data in multiple returns. We are unsure whether Leverage will become part of the COREP DPM, or whether it will have a separate DPM for itself. If it aligns under the COREP model, the fact that the LR6 template is duplicating much of the COREP credit template begs the questions as to why would the LR6 template be needed at all if the data already exists within the DPM submitted. Thus, we would like to know what the extra data beyond that required for the pure LR calculations will be used for. Also, we would ask that consideration be given to minimising the duplication of data requests for the LR templates if the data is already requested in the COREP templates 2. Overall, at this stage, we also feel that the quality of the instructions in the templates could be improved. Please see Appendix 1. Q2: Do institutions already have the data required under this proposal on a monthly basis? If so, is this data of the required standard as other data reported to supervisory authorities? As described above, there are complexities with the additional data items that are not currently reported and with data classification. For data which is superfluous to the LR calculation, this data should be provided on a quarterly basis at a minimum if it is definitely required by the EBA. 2 We are aware that in some circumstances, it may be that the leverage reporting is not wholly consistent with capital reporting. For example, banks which have approval n place to use internal models for derivatives and secured financing transactions (SFTs) could be required to apply the standardised modelled calculations for leverage which would differ form the capital treatment and result in additional calculations and reporting capabilities representing significant challenge. Additionally, standardised conversion factors are required for off-balance sheet items even though capital reporting uses advanced conversion factors. Page 4 of 10

6 AFME and its members are concerned about the different frequencies of quarterly COREP reporting versus gathering monthly LR data that is cross-referenced to the COREP templates. The industry supports CRR Article 475 (3) that gives derogation to competent authorities to allow institutions to report end-of-quarter data to AFME members support the use of quarterly averages as it is impossible for some institutions to provide the calculations on a monthly basis. An alternative option would be to do a point-in-time calculation at quarter-end as there are many cross-references to other COREP templates which are only produced at quarter-end. This would be the favoured option as it is reconcilable directly with COREP for better consistency amongst supervisory reporting. Q3: The same timelines are proposed for reporting on a consolidated level as well as on an individual level, is this seen as problematic? If so, would you propose a different timeline for reporting on a consolidated level? The proposal for multiple levels (consolidated, sub-group, standalone) of reporting does not fully correspond with the majority of the current statutory or regulatory reporting requirements. Generally, the existing reporting systems at banks may not be capable of producing the sub-group level data. This is as when combining the legal entity level data, netting has to be performed on all transactions between the sub-group entities. Therefore, significant manual intervention is required until automated sub-group reporting capabilities are built. AFME proposes an added business days for consolidated submissions to allow for inter sub-group transaction netting to take place after the standalone reports are compiled. Additionally, institutions are not sure whether reporting on an individual basis means reporting at legal entity or solo level basis. In practice, the reporting would be only for significant legal entities. Further guidance is sought on this issue not only for the LR reports for the COREP reports as well. In principle, we would encourage the alignment between COREP and the LR templates in this respect. Q4: What additional costs do you envisage from the proposed approach to reporting the leverage ratio in order to fulfil the requirements of the CRR outlined in this ITS? The additional data items, not covered in other areas of regulatory or statutory reporting and not necessary for the provision of the LR, will add substantial cost as firms may not have the data readily available. Also, the proposed templates require multiple methods of derivative add-on and SFT netting to be recorded. As only one method feeds into the leverage calculation, the extra requirements could result in superfluous data collection and analysis. The cost is firm-specific as building the technical reporting capabilities depend on firm-specific reporting tools and data capture methods. The consequent manual effort will be allocated to the operational staff that already deals with increased regulatory reporting requirements. AFME is concerned that any duplication of data requirements and differences in classification between templates can add substantially to the reporting burden, if the reporting requirements are not carefully considered and aligned where Page 5 of 10

7 possible. The impact of the excess burdens on firms is likely to be reflected in the quality of the data submitted which may be below par initially. The costs associated with the EBA requirement for sub-group/legal entity level reporting have two streams. Firstly, the technological changes required to facilitate non-statutory level reporting will add to the already high IT infrastructure costs. Secondly, carving out the inter-sub-group transactions to achieve the correct reporting group exposures will be performed manually. Thus, this strain on the reporting staff increases substantially over key periods. Also, we would request that any additional data items that are not linked directly to the leverage ratio calculations should only be collected on a quarter-end basis. This is as they require new IT builds that are subject to systems and data quality testing, additional operational risk and additional burden on the regulatory and finance resources. Questions from Annex II: Q5: Is the calculation of the derivatives share threshold sufficiently clear? AFME and its members request more clarity on the thresholds, especially for firms that provide agency type services to clients and how these exposures are treated in identifying the relevant exposures. AFME members believe that examples of how the calculation applies to different activities would help in determining the appropriate calculation methods. Q6: Do you believe this method captures institutions derivatives exposure in a sensible way? AFME notes that netting of risk is not captured in the template. Firms are unable to offset the exposures with various types of collateral that they may hold against the exposures. This appears to be disproportionate to businesses that manage risk based on a portfolio model, whereby exposures are netted across multiple exposures and managed on a net basis. Q7: Does the reduction of fields to be reported in a given period by institutions that do not exceed the threshold value in that period, lead to a significant reduction in administrative burden? We do not have a view on this; we believe most AFME members exposures exceed the threshold. Q8: Preliminary internal calculations by supervisors suggest that a threshold value should be in the range of 0.5% to 2%. Would you suggest a different threshold level, if yes, please justify this? Page 6 of 10

8 We do not have a view on this; we believe most members exposures exceed the threshold. We are not clear on the relationship between the two thresholds. Please also see response to Q10. Q9: Is the calculation of the nominal amount threshold sufficiently clear? This is an additional data item that is not required for the calculation of the LR. Furthermore, this data is not required for any current reporting regulatory or statutory reporting. AFME and its members would like to know the drivers for collecting this data to understand what it is used for. If intrinsic benefits from collecting this data cannot be explained, AFME recommends excluding the LR4 items from the leverage ratio template to avoid unfounded burdensome data collection. At the very least, collection if such data items should be delayed. Q10: Preliminary internal calculations by supervisors suggest that the nominal threshold value should be in the range of 200 to 500 million.. Would you suggest a different threshold level, if yes, please justify this? As a credit derivative is a derivative, we are uncertain as to how the two thresholds relate to each other. For example, if an entity does not reach the threshold to require derivatives reporting, but does reach the threshold for credit derivatives should it report the credit derivatives piece? We are not clear on this issue. Q11: Is the term reference name and the distinction from reference obligation sufficiently clear? We understand reference name to refer to the underlying legal entity and thereby to be broader in scope than reference obligation which we interpret to refer to specific obligations of the reference name. We would, nonetheless, welcome further clarification on this. Q12: Is the treatment of credit derivatives referring to indices and baskets sufficiently clear? The treatment of nth-to-default and tranched credit derivative transactions is unclear in LR4. We would welcome clarity as to whether they should be included or excluded. Q13: Which additional contractual features should be taken into consideration when assessing offsetting of written and purchased credit derivatives? How would this add to complexity and reporting burden? AFME and its members advocate the EBA to consider the different purposes for holding such instruments in the banking and trading books and what needs to be reported in the supplementary fields of the LR. Page 7 of 10

9 Q14: Is the classification used in template LR6 sufficiently clear? AFME and its members have noted that the exposure classes are the same as in the COREP credit template. To avoid double counting and additional complexities in gathering the data, the classifications in the templates should be aligned, or the LR6 abolished as the same data is reported for credit risk already. Please also see the response to Q1. Q15: Do you believe the current split, which is predominantly based on the exposure classes for institutions using the standard method are appropriate or would you suggest an alternative split? To avoid duplication and artificial differences between data items in the different reporting templates, we recommend that the data split differences should be kept at minimum levels. Doing this would allow for greater transparency of the data and minimise reporting differences that are caused by interpretation of the technical standards. In fact, we would support that any duplicate data items be abolished altogether if at all possible. Q16: Is the classification used in template LR7 sufficiently clear? The classifications used in the template are clear, apart from the sub-group consolidations. AFME and its members would benefit from further clarity on the sub-group reporting requirements. Conclusion We would be pleased, of course, to discuss the issues covered in this submission with the EBA or to provide further information about any of the matters which our members have raised if that would be helpful. AFME 24 August 2012 Page 8 of 10

10 Appendix 1 Suggestions for improvements on the instructions for the templates LR 6 Row there are no corresponding instructions The formulae on page 3 do not work: referencing non-existent cells and greyed out cells, etc. The formula on Annex II, page 3, reference 13: {LR 2;070;5} is referred to three times in the formula: a) one is * {LR2;070;5} b) the second is - {LR2;070;5} c) the third is - {LR2;070;5} Is this what was intended? The thresholds have not been included The CP and instructions don't actually give the threshold values - just big X's (see Annex II, page 3, reference 24) The definitions are very poor What is an accounting balance sheet value (i.e., using the firms accounting standards?) The instructions are incomplete and where provided unclear For example, there are no clear instructions on signage: so for example - CDS protection sold versus bought. There is much duplication if data items requested Credit derivatives appear on LR 1, LR 2, LR 4. Are these overlaps in data or is there a difference between the derivatives captured in each return? Specific Comments on Templates o LR2 From the guidance it is not clear what the difference between Method 1 and Method 2 is. Would they be the same under ISDA? Where would we put the default funds? Would the funded piece will go into the Other Assets in LR1, and the unfunded piece in the Off balance sheet items with a 100% CCF in the RSA? o LR4 This report is unclear. Is it suggesting that we report at reference name level or just one high level number? Page 9 of 10

11 Column 3 is unclear, especially the text around where the reporting institution is selling credit protection on the same underlying reference entity. Would interpret this as the net? The following guidance is also unclear For each reference name, the notional amounts of credit protection bought which are considered in this field must not exceed the notional amounts of the credit protection. Is this a new limit? If not and Bought is 100 and Sold is 80, would we report Bought as 80 (even though this is inaccurate as it is actually 100)? o LR6 Is row 020 the cash side of the SFT and row 030 the security side? Or is 020 both and 030 just the security? From row 80 down is it all non trading book? Page 10 of 10